d_hadzima

How does VB handle dividion by zero. I am using it to trap invalid data and I am wondering if division by zero is going to slow the perfomance. Here is just an example.
Thanks,
D

Private Sub cmdUpdate_Click()
On Error GoTo divZero
Dim i As Integer
Dim j As Integer
Dim projectLine As String

If (lstforms.ListCount / lstProjects.ListCount) > 0 Then

For j = 1 To lstProjects.ListCount
For i = 1 To lstforms.ListCount
lstforms.ListIndex = i - 1
projectLine = BuildProjectLine(lstforms.Text)
Debug.Print projectLine
Next i
Next j
End If

Exit Sub
divZero:
MsgBox "Please enter all data correctly.", vbInformation, "Learn How To Work This"

End Sub

Flyguy

I would do it like this:

BillSoo

It should not slow performance because you are only using it once, at the top of your loop. Error handling *is* a bit slow, but in your case it should not matter.

A larger problem is that your error handler will catch *any* unhandled error, not just div by zero, including those in your function BuildProjectLine...which can lead to unexpected results.

I would just have a simple

If lstProjects.ListCount Then

at the beginning to pass only a non-zero listcount. It's simple and more readable.

d_hadzima

Thanks Arn. I understand how to do this but what I was really looking for are the ramifications of using division by zero to trap errors/invalid data. Does it hurt/better the performance of a program.
Thanks again for the response.
D

d_hadzima

Too late on that one. Thanks Bill that is what I was looking for.
D

BillSoo

As I said, an error trap will trap ALL errors in scope, not just div by zero. So you then have to sort them out.

It's as if all errors were tossed into a big pile and your error handler then has to sort through them and figure out which came where.

For instance, suppose you had 2 places where a div by zero error could occur. How would you know which one actually triggered the error? How could you then fix it?

Flyguy

Even worse your code contained multiple errors, which will raise an error in any given case.
I marked the correct code in bold. So don't say you know how to do this.

d_hadzima

I undertand all of this. The example I used was merely an attempt to better show what knowledge I was trying to obtain.

I was planning on using this for interpolating Taylor Polynomials. Eventually if I keep interpolating my error will come infinately close to zero. I wan't it to handle it "nicely" if I hit zero.

Thanks for the help though guys.
D

d_hadzima

Just because I don't use a list does not mean it is going to throw an error. So I don't know where you are getting multiple errors from.

BillSoo

The usual way to handle this situation is to have test for a value below a certain value. Otherwise, very small,yet non-zero values can trigger overflow errors when you try to divide by them....

d_hadzima

Never Mind Arn, I see what you mean now. I never ran this I just typed it in because I never had any vision that I would use it. Sorry I hit the fencepost on you.
Have a good weekend all.
D